An algorithmic approach to building dynamic ratings in organizational sports management systems

The relevance of this study is driven by the growing need for objective and adaptive methods to assess team strength within the dynamically changing environment of sports management, where traditional rating systems inadequately account for the temporal decay of results, home advantage, and league-specific context. Accordingly, this article aims to develop an algorithmic approach to constructing dynamic sports team ratings, thereby improving forecast accuracy and ensuring assessment reliability. The core methodology employs an original iterative algorithm based on two-way recalculation of performance indicators, utilizing nonlinear weighting coefficients and a temporal decay factor to comprehensively evaluate how historical results and time dynamics influence current team competitiveness. The article presents the mathematical model and rating computation procedure, elucidates the mechanisms of two-stage baseline stabilization and dynamic value calculation, and establishes a quantitative relationship between prediction accuracy and the decay parameter: an optimal value of λ = 0.993 yields a correct prediction rate of 67.419%, representing a 5.539 percentage-point increase over the baseline model without temporal adjustment (61.880%). The algorithm’s stability within the operational range of λ ∈ [0.989; 0.995] is also substantiated. These findings offer practical value for sports management professionals, club analytics departments, and competition organizers, as the developed model, implemented as a software module, can support managerial decision-making in match forecasting, schedule optimization, and strategic planning for sports organizations.

1. Yushkin V.N., Marchenko S.S., Strizhakova E.A., Penkova R.I. Comparative analysis of methods for predicting the results of competitive activity. Vestnik sportivnoi nauki. 2023;(6):25–31. (In Russ.).

2. Miazek A., Podleśny A. Methods of creating sports rankings based on selected examples. Sport and Tourism Central European Journal. 2021;4(3):75–103. (In Polish). https://doi.org/10.16926/sit.2021.04.19

3. Yevlanov M., Vasyltcova N., Uvarova V. Development of a method of forecasting the results of the performances of athletes and teams in sports competitions. Automated Control Systems and Automation Devices. 2024;(180):72–88. https://doi.org/10.30837/0135-1710.2024.180.072

4. Grudtsyna L.Yu., Akhmetzyanov R.F. Forecasting the results of sports competitions based on machine learning methods. Scientific and Technical Volga region Bulletin. 2024;(5):92–95. (In Russ.).

5. Elo A.E. The Rating of Chessplayers: Past and Present. Bronx: Ishi Press International; 2008. 208 p.

6. Grinev I.V., Usmanov S.R., Paramonov I.N. Selection of ELO rating parameters in armwrestling. Scientific notes of P.F. Lesgaft University. 2023;(3):118–120. (In Russ.).

7. Glickman M.E. Parameter estimation in large dynamic paired comparison experiments. Journal of the Royal Statistical Society Series C: Applied Statistics. 1999;48(3):377–394. https://doi.org/10.1111/1467-9876.00159

8. Herbrich R., Minka T., Graepel Th. TrueSkill™: A Bayesian skill rating system. In: Advances in Neural Information Processing Systems 19: Proceedings of the Twentieth Annual Conference on Neural Information Processing Systems, 04–07 December 2006, Vancouver, Canada. Cambridge: MIT Press; 2007. P. 569–576. https://doi.org/10.7551/mitpress/7503.003.0076

9. Baio G., Blangiardo M. Bayesian hierarchical model for the prediction of football results. Journal of Applied Statistics. 2010;37(2):253–264. https://doi.org/10.1080/02664760802684177

10. Cattelan M., Varin C., Firth D. Dynamic Bradley-Terry modelling of sports tournaments. Journal of the Royal Statistical Society Series C: Applied Statistics. 2013;62(1):135–150. https://doi.org/10.1111/j.1467-9876.2012.01046.x

11. Knorr-Held L. Dynamic rating of sports teams. Journal of the Royal Statistical Society Series D: The Statistician. 2000;49(2):261–276. https://doi.org/10.1111/1467-9884.00236

12. Krutikov A.K., Klyukin V.L., Podkovyrin V.D. Solving the problem of predicting the results of individual sports disciplines using a cascade of neural networks. Scientific Review. Technical Science. 2022;(3):10–14. (In Russ.). https://doi.org/10.17513/srts.1393

13. Teryohin A.D., Ilyalov O.R., Stepanov A.V. Assessment system for sports exercises by neural network video analysis. Applied Mathematics and Control Sciences. 2022;(1):75–86. (In Russ.).

14. Butina D.V., Yasnitsky L.N. Neural network system for predicting the Italian football lega "Serie A" season results. Bulletin of Perm University. Mathematics. Mechanics. Computer Science. 2023;(1):84–92. (In Russ.). https://doi.org/10.17072/1993-0550-2023-1-84-92

15. Simonova M.M., Udalov D.E., Butyrina S.A. Features of project management in the sports industry. Problemy teorii i praktiki upravleniia. 2020;(1):35–47. (In Russ.).

16. Esin S.A. Effective management decisions as a factor in the development of organizations in the sports industry. Scientific Notes of the Russian Academy of Entrepreneurship. 2023;22(4):95–99. (In Russ.). https://doi.org/10.24182/2073-6258-2023-22-4-95-99

17. Barros C.P., Leach S. Technical efficiency in the English Football Association Premier League with a stochastic cost frontier. Applied Economics Letters. 2007;14(10):731–741. https://doi.org/10.1080/13504850600592440

18. Glickman M.E. Example of the Glicko-2 system. Boston University. 2012;28:2012.

19. Hunter D.R. MM algorithms for generalized Bradley-Terry models. The Annals of Statistics. 2004;32(1):384–406. https://doi.org/10.1214/aos/1079120141